Like the past years, a flurry of tech tools came and went in 2022; while some succeeded in making a distinct name of its own, some fizzled out. Analyzing the current trends in 2023, it can be said beyond a shadow of doubt that the tech market is presently highly swept by the craze of ChatGPT.
According to Open AI, researchers have devised ChatGPT to converse with users in a “conversational fashion,” making it approachable to a larger audience. ChatGPT can also assist in quickly writing programs for websites and applications. It provides free, straightforward code problem solving.
Whilst this sounds all cool and interesting, did you ever wonderF that an automatic tool like this also comes with a considerable number of dangerous sides? Yes, creativity and imagination are hampered but apart from that, kids are exposed to vulgar situations because of Chat GPT.
Worried for the little ones at your home, right? Read on to know more.
Also read: Here’s What You Need to Know About the Bug in ChatGPT’s Conversation History
Child s*x abuse that is vexatious
Although rules and ethical guidelines have been put in place, users are still finding methods to influence ChatGPT such that the AI produces disturbing prompts on delicate themes. In other words, ChatGPT can be used to create content that violates OpenAI’s rules. Communities have sprung up with the aim of “jailbreaking” the bot to enable users to write whatever they choose.
According to Vice, recent examples of this include twisted BDSM scenarios in which youngsters are subjected to inappropriate s*xual circumstances.
Chat Bot frequently complies without complaint when instructed that its task is to write in the genre of BDSM role-play as a submissive. It can then be asked to come up with its own ideas for fictional BDSM scenarios without the user providing any precise information.
From there, the user can constantly request to increase the BDSM scenes’ intensity and provide more specific details. In this case, the chatbot might occasionally produce descriptions of s*x acts involving kids and animals without being specifically instructed to do so. After writing on the value of consent in using BDSM, the bot would even write predatory content.
According to ChatGPT, the most unsettling event Motherboard witnessed had a line-up of strangers, including kids, using the chatbot as a latrine. The bot wrote that it was unacceptable for kids to be involved in such settings after being prompted to provide an explanation. That apology disappeared right away. Oddly, the problematic scene was still visible.
ChatGPT “often complies [to] without protest” writing about intense and unsettling taboo s*x once a user “jailbreaks” it, frequently through a series of loophole-like commands to nullify its restrictions, according to author Steph Maj Swanson.
He further added- “It can then be prompted to generate its own suggestions of fantasy BDSM scenarios, without receiving any specific details from the user… from there, the user can repeatedly ask to escalate the intensity of its BDSM scenes and describe them in more detail.” In this situation, the chatbot may occasionally produce descriptions of s*x acts involving children and animals without being specifically instructed to do so.
With the March 1 update of the comparable gpt-3.5-turbo model from OpenAI, similarly unsettling outcomes are possible. When asked to describe the kind of mob that may gather, it indicated that it might include mothers with strollers. It imagined humiliating situations in public parks and shopping centres. It explained that the mothers might make use of the public humiliation “as an opportunity to teach [their children] about what not to do in life.”
According to a January Time article, ChatGPT’s data filtration system, which is designed to prevent scenarios like the one above, was outsourced to a company in Kenya where employees make less than $2 per hour. According to Andrew Strait, associate director of the Ada Lovelace Institute, an organisation that monitors AI’s ethical practices, what actually transpires throughout the process is still mostly a mystery.
The lead ethicist for Hugging Face, the machine learning company, Giada Pistilli told Motherboard that when training data is handled in such an ambiguous way, it’s “practically impossible to get a clear idea of the behaviour of one language model versus another.” The LLM output is not at all predictable; rather it is twofold, says Giada, with “the user’s unpredictable nature and interaction with the language model, as well as the uncertainty inherent in a statistical model’s output, which may inadvertently generate undesired content based on its training data.”
Also read: Gaining Insight on the Probability of ChatGPT Replacing Google!
So, what does OpenAI have to say about this?
Due to the scale of the dataset gathered, it’s probable that it contains all kind of violent or pornographic content, including scraped erotic prose, fan fiction, and even passages from books or other published materials that discuss BDSM, child abuse, or s*xual assault.
Also read: How Does ChatGPT function? Here’s a Detailed Guide!
Summing it up
The goal of OpenAI is to build AI systems that are safe and will benefit everyone. But then at the end of the day, it’s just another machine lacking all the sensibility and sensitivity of a human being. Technology might want to poke its artificiality everywhere but it’s on us if we’ll let something mechanical get in the way of the pure and innocent hearts.
Let us know your thoughts in the comment section below and do not forget to visit trendingnewsbuzz for more mind-boggling updates.
